
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from scipy.stats import shapiro 

from statsmodels.stats.outliers_influence import variance_inflation_factor 
from sklearn.model_selection import train_test_split, GridSearchCV, RandomizedSearchCV

from sklearn.linear_model import LogisticRegression
from sklearn.tree import plot_tree, DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor, AdaBoostRegressor, GradientBoostingRegressor
from sklearn.metrics import mean_absolute_error,mean_squared_error,r2_score

from sklearn.svm import SVC


df=pd.read_csv(r"D:\Excel project\energy_dataset_.csv")
df


df.size

195000


df.shape

(15000, 13)


df.axes

[RangeIndex(start=0, stop=15000, step=1),
 Index(['Type_of_Renewable_Energy', 'Installed_Capacity_MW',
        'Energy_Production_MWh', 'Energy_Consumption_MWh',
        'Energy_Storage_Capacity_MWh', 'Storage_Efficiency_Percentage',
        'Grid_Integration_Level', 'Initial_Investment_USD', 'Funding_Sources',
        'Financial_Incentives_USD', 'GHG_Emission_Reduction_tCO2e',
        'Air_Pollution_Reduction_Index', 'Jobs_Created'],
       dtype='object')]


df.columns

Index(['Type_of_Renewable_Energy', 'Installed_Capacity_MW',
       'Energy_Production_MWh', 'Energy_Consumption_MWh',
       'Energy_Storage_Capacity_MWh', 'Storage_Efficiency_Percentage',
       'Grid_Integration_Level', 'Initial_Investment_USD', 'Funding_Sources',
       'Financial_Incentives_USD', 'GHG_Emission_Reduction_tCO2e',
       'Air_Pollution_Reduction_Index', 'Jobs_Created'],
      dtype='object')


df.describe()


df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 15000 entries, 0 to 14999
Data columns (total 13 columns):
 #   Column                         Non-Null Count  Dtype  
---  ------                         --------------  -----  
 0   Type_of_Renewable_Energy       15000 non-null  int64  
 1   Installed_Capacity_MW          15000 non-null  float64
 2   Energy_Production_MWh          15000 non-null  float64
 3   Energy_Consumption_MWh         15000 non-null  float64
 4   Energy_Storage_Capacity_MWh    15000 non-null  float64
 5   Storage_Efficiency_Percentage  15000 non-null  float64
 6   Grid_Integration_Level         15000 non-null  int64  
 7   Initial_Investment_USD         15000 non-null  float64
 8   Funding_Sources                15000 non-null  int64  
 9   Financial_Incentives_USD       15000 non-null  float64
 10  GHG_Emission_Reduction_tCO2e   15000 non-null  float64
 11  Air_Pollution_Reduction_Index  15000 non-null  float64
 12  Jobs_Created                   15000 non-null  int64  
dtypes: float64(9), int64(4)
memory usage: 1.5 MB


df.isna().sum()

Type_of_Renewable_Energy         0
Installed_Capacity_MW            0
Energy_Production_MWh            0
Energy_Consumption_MWh           0
Energy_Storage_Capacity_MWh      0
Storage_Efficiency_Percentage    0
Grid_Integration_Level           0
Initial_Investment_USD           0
Funding_Sources                  0
Financial_Incentives_USD         0
GHG_Emission_Reduction_tCO2e     0
Air_Pollution_Reduction_Index    0
Jobs_Created                     0
dtype: int64


df.dtypes

Type_of_Renewable_Energy           int64
Installed_Capacity_MW            float64
Energy_Production_MWh            float64
Energy_Consumption_MWh           float64
Energy_Storage_Capacity_MWh      float64
Storage_Efficiency_Percentage    float64
Grid_Integration_Level             int64
Initial_Investment_USD           float64
Funding_Sources                    int64
Financial_Incentives_USD         float64
GHG_Emission_Reduction_tCO2e     float64
Air_Pollution_Reduction_Index    float64
Jobs_Created                       int64
dtype: object


plt.figure(figsize=(10, 6))
sns.heatmap(df.corr(),annot=True,cmap='rainbow')

<Axes: >


df.var()

Type_of_Renewable_Energy         3.997520e+00
Installed_Capacity_MW            8.306666e+04
Energy_Production_MWh            2.075412e+10
Energy_Consumption_MWh           1.669876e+10
Energy_Storage_Capacity_MWh      8.375794e+06
Storage_Efficiency_Percentage    2.098236e+02
Grid_Integration_Level           1.261816e+00
Initial_Investment_USD           2.053038e+16
Funding_Sources                  6.682952e-01
Financial_Incentives_USD         3.349287e+13
GHG_Emission_Reduction_tCO2e     2.067532e+08
Air_Pollution_Reduction_Index    8.154781e+02
Jobs_Created                     2.106018e+06
dtype: float64


df.cov()


df.std()

Type_of_Renewable_Energy         1.999380e+00
Installed_Capacity_MW            2.882129e+02
Energy_Production_MWh            1.440629e+05
Energy_Consumption_MWh           1.292237e+05
Energy_Storage_Capacity_MWh      2.894096e+03
Storage_Efficiency_Percentage    1.448529e+01
Grid_Integration_Level           1.123306e+00
Initial_Investment_USD           1.432843e+08
Funding_Sources                  8.174933e-01
Financial_Incentives_USD         5.787303e+06
GHG_Emission_Reduction_tCO2e     1.437892e+04
Air_Pollution_Reduction_Index    2.855658e+01
Jobs_Created                     1.451213e+03
dtype: float64


sns.pairplot(df)
plt.title('Jobs_Created')
plt.show()


plt.scatter(df['Type_of_Renewable_Energy'], df['Jobs_Created'])
plt.show()


sns.boxplot(df["Type_of_Renewable_Energy"])

<Axes: >


sns.boxplot(df["Installed_Capacity_MW"])

<Axes: >


sns.boxplot(df["Installed_Capacity_MW"])

<Axes: >


sns.boxplot(df["Energy_Production_MWh"])

<Axes: >


sns.boxplot(df["Energy_Consumption_MWh"])

<Axes: >


sns.boxplot(df["Energy_Storage_Capacity_MWh"])

<Axes: >


sns.boxplot(df["Storage_Efficiency_Percentage"])

<Axes: >


sns.boxplot(df["Grid_Integration_Level"])

<Axes: >


sns.boxplot(df["Initial_Investment_USD"])

<Axes: >


sns.boxplot(df["Funding_Sources"])

<Axes: >


sns.boxplot(df["Financial_Incentives_USD"])

<Axes: >


df.shape

(15000, 13)


df1 = df.iloc[:, :12]
df1


vif_df = pd.DataFrame()

vif_df["independant Features"] = df1.columns

vif_df


vif_list = []


for i in range(df1.shape[1]):
    vif = variance_inflation_factor(df1.to_numpy(),i)
    vif_list.append(vif)


vif_df["VIF"] = vif_list
vif_df


df.shape

(15000, 13)


x = df1         
y = df["Jobs_Created"]


xtrain,xtest,ytrain,ytest = train_test_split(x, y, test_size=0.3, random_state=11)


xtrain.shape

(10500, 12)


ytrain.shape

(10500,)


xtest.shape

(4500, 12)


ytest.shape

(4500,)


log_reg = LogisticRegression()
log_reg

LogisticRegression()

LogisticRegression()


log_reg_model = log_reg.fit(xtrain, ytrain)
log_reg_model

C:\Users\prajw\anaconda3\lib\site-packages\sklearn\linear_model\_logistic.py:458: ConvergenceWarning: lbfgs failed to converge (status=1):
STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.

Increase the number of iterations (max_iter) or scale the data as shown in:
    https://scikit-learn.org/stable/modules/preprocessing.html
Please also refer to the documentation for alternative solver options:
    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression
  n_iter_i = _check_optimize_result(

LogisticRegression()

LogisticRegression()


ytrain.head(10)

14314    4050
2415     2670
8169     1737
12199    1631
2737     3030
12032    3382
12452     148
861      4120
4661     2168
8245     4360
Name: Jobs_Created, dtype: int64


ytrain_pred = log_reg_model.predict(xtrain)
ytrain_pred

array([1045, 4888, 1045, ..., 1045, 4054, 3470], dtype=int64)


ytrain_pred = log_reg_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 3796829.2994285715
mean absolute error : 1588.490476190476
Root mean squared error : 1948.5454317076037
R2 score: -0.7948788777442457


ytest_pred = log_reg_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 3812043.5486666667
mean absolute error : 1594.8197777777777
Root mean squared error : 1952.4455302688132
R2 score: -0.830701998967226


dt_reg = DecisionTreeRegressor()
dt_reg

DecisionTreeRegressor()

DecisionTreeRegressor()


dt_reg_model = dt_reg.fit(xtrain,ytrain)
dt_reg_model

DecisionTreeRegressor()

DecisionTreeRegressor()


ytrain_pred = dt_reg_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 0.0
mean absolute error : 0.0
Root mean squared error : 0.0
R2 score: 1.0


ytest_pred = dt_reg_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 4294379.726666667
mean absolute error : 1696.8528888888889
Root mean squared error : 2072.288523991451
R2 score: -1.0623399102255227


hyperparameters = {
    "criterion" : ["squared_error", "absolute_error", "friedman_mse", "poisson"],
    "max_depth" : np.arange(2,50),
    "min_samples_split": np.arange(2,10),
    "min_samples_leaf" : np.arange(2,15)
}
print(hyperparameters)

{'criterion': ['squared_error', 'absolute_error', 'friedman_mse', 'poisson'], 'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]), 'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9]), 'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14])}


rscv = RandomizedSearchCV(dt_reg_model, hyperparameters, cv=5) #cv : cross validation
rscv

RandomizedSearchCV(cv=5, estimator=DecisionTreeRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9])})

RandomizedSearchCV(cv=5, estimator=DecisionTreeRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9])})

DecisionTreeRegressor()

DecisionTreeRegressor()


rscv_reg = rscv.fit(xtrain, ytrain)
rscv_reg

RandomizedSearchCV(cv=5, estimator=DecisionTreeRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9])})

RandomizedSearchCV(cv=5, estimator=DecisionTreeRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9])})

DecisionTreeRegressor()

DecisionTreeRegressor()


rscv_reg.best_estimator_

DecisionTreeRegressor(max_depth=3, min_samples_leaf=7, min_samples_split=7)

DecisionTreeRegressor(max_depth=3, min_samples_leaf=7, min_samples_split=7)


rscv_reg_hyp = DecisionTreeRegressor(criterion='poisson', max_depth=10, min_samples_leaf=3,
                      min_samples_split=7)
rscv_reg_hyp

DecisionTreeRegressor(criterion='poisson', max_depth=10, min_samples_leaf=3,
                      min_samples_split=7)

DecisionTreeRegressor(criterion='poisson', max_depth=10, min_samples_leaf=3,
                      min_samples_split=7)


dt_reg_hyp_model = rscv_reg_hyp.fit(xtrain,ytrain)
dt_reg_hyp_model

DecisionTreeRegressor(criterion='poisson', max_depth=10, min_samples_leaf=3,
                      min_samples_split=7)

DecisionTreeRegressor(criterion='poisson', max_depth=10, min_samples_leaf=3,
                      min_samples_split=7)


ytrain_pred = dt_reg_hyp_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 1922712.719504361
mean absolute error : 1177.1810715617278
Root mean squared error : 1386.619168879603
R2 score: 0.09107411051427561


ytest_pred = dt_reg_hyp_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2294414.3971356065
mean absolute error : 1297.390343694187
Root mean squared error : 1514.7324506775467
R2 score: -0.10187330487462609


plot_tree(dt_reg_hyp_model, feature_names=df.columns,class_names= list("ABCDEF"), filled=True)
plt.savefig("dt_hyp_model.png")


rf_reg = RandomForestRegressor()
rf_reg

RandomForestRegressor()

RandomForestRegressor()


rf_reg_model = rf_reg.fit(xtrain,ytrain)
rf_reg_model

RandomForestRegressor()

RandomForestRegressor()


ytrain_pred = rf_reg_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 300817.0957437714
mean absolute error : 468.8258590476191
Root mean squared error : 548.4679532513923
R2 score: 0.8577944361901852


ytest_pred = rf_reg_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2141911.119747
mean absolute error : 1265.6518688888889
Root mean squared error : 1463.526945343679
R2 score: -0.028634882700244946


hyperparameters = {
    "n_estimators" : np.arange(2,50),
    "criterion" : ["squared_error", "absolute_error", "friedman_mse", "poisson"],
    "max_depth" : np.arange(2,50),
    "min_samples_split": np.arange(2,10),
    "min_samples_leaf" : np.arange(2,15)
}
print(hyperparameters)

{'n_estimators': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]), 'criterion': ['squared_error', 'absolute_error', 'friedman_mse', 'poisson'], 'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]), 'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9]), 'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14])}


rscv_rf = RandomizedSearchCV(rf_reg_model, hyperparameters, cv=5)
rscv_rf

RandomizedSearchCV(cv=5, estimator=RandomForestRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9]),
                                        'n_estimators': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49])})

RandomizedSearchCV(cv=5, estimator=RandomForestRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9]),
                                        'n_estimators': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49])})

RandomForestRegressor()

RandomForestRegressor()


rscv_rf_reg = rscv_reg.fit(xtrain,ytrain)
rscv_rf_reg

RandomizedSearchCV(cv=5, estimator=DecisionTreeRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9])})

RandomizedSearchCV(cv=5, estimator=DecisionTreeRegressor(),
                   param_distributions={'criterion': ['squared_error',
                                                      'absolute_error',
                                                      'friedman_mse',
                                                      'poisson'],
                                        'max_depth': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
       19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
       36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]),
                                        'min_samples_leaf': array([ 2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
                                        'min_samples_split': array([2, 3, 4, 5, 6, 7, 8, 9])})

DecisionTreeRegressor()

DecisionTreeRegressor()


rscv_rf_reg.best_estimator_

DecisionTreeRegressor(criterion='friedman_mse', max_depth=7, min_samples_leaf=3,
                      min_samples_split=8)

DecisionTreeRegressor(criterion='friedman_mse', max_depth=7, min_samples_leaf=3,
                      min_samples_split=8)


rscv_rf_reg = DecisionTreeRegressor(criterion='poisson', max_depth=3, min_samples_leaf=8,
                      min_samples_split=8)
rscv_rf_reg

DecisionTreeRegressor(criterion='poisson', max_depth=3, min_samples_leaf=8,
                      min_samples_split=8)

DecisionTreeRegressor(criterion='poisson', max_depth=3, min_samples_leaf=8,
                      min_samples_split=8)


rscv_rf_reg_model = rscv_rf_reg.fit(xtrain,ytrain)


ytrain_pred = rscv_rf_reg_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 2103156.847674949
mean absolute error : 1255.1765315363575
Root mean squared error : 1450.2264815107153
R2 score: 0.005772578966594999


ytest_pred = rscv_rf_reg_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2090347.9410888588
mean absolute error : 1251.2156414210908
Root mean squared error : 1445.8035624139466
R2 score: -0.0038720978481199264


adb_reg = AdaBoostRegressor()
adb_reg_model = adb_reg.fit(xtrain,ytrain)
adb_reg_model

AdaBoostRegressor()

AdaBoostRegressor()


ytrain_pred = adb_reg_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 2111358.455553671
mean absolute error : 1259.2899533595385
Root mean squared error : 1453.051429080771
R2 score: 0.0018954247454995299


ytest_pred = adb_reg_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2083499.1601479966
mean absolute error : 1250.8864015602019
Root mean squared error : 1443.4331159246683
R2 score: -0.0005830281407945836


hyp = {"n_estimators" : np.arange(2,100),
       "learning_rate" : [0, 0.1, 0.001, 0.0001, 1]}


RandomizedSearchCV(adb_reg_model, hyp, cv=5).fit(xtrain,ytrain).best_estimator_

C:\Users\prajw\anaconda3\lib\site-packages\sklearn\model_selection\_validation.py:378: FitFailedWarning: 
10 fits failed out of a total of 50.
The score on these train-test partitions for these parameters will be set to nan.
If these failures are not expected, you can try to debug them by setting error_score='raise'.

Below are more details about the failures:
--------------------------------------------------------------------------------
10 fits failed with the following error:
Traceback (most recent call last):
  File "C:\Users\prajw\anaconda3\lib\site-packages\sklearn\model_selection\_validation.py", line 686, in _fit_and_score
    estimator.fit(X_train, y_train, **fit_params)
  File "C:\Users\prajw\anaconda3\lib\site-packages\sklearn\ensemble\_weight_boosting.py", line 124, in fit
    self._validate_params()
  File "C:\Users\prajw\anaconda3\lib\site-packages\sklearn\base.py", line 581, in _validate_params
    validate_parameter_constraints(
  File "C:\Users\prajw\anaconda3\lib\site-packages\sklearn\utils\_param_validation.py", line 97, in validate_parameter_constraints
    raise InvalidParameterError(
sklearn.utils._param_validation.InvalidParameterError: The 'learning_rate' parameter of AdaBoostRegressor must be a float in the range (0, inf). Got 0 instead.

  warnings.warn(some_fits_failed_message, FitFailedWarning)
C:\Users\prajw\anaconda3\lib\site-packages\sklearn\model_selection\_search.py:952: UserWarning: One or more of the test scores are non-finite: [-0.0017325  -0.00123039 -0.0015858  -0.00018256         nan -0.000965
         nan -0.00167245 -0.00123635 -0.0010247 ]
  warnings.warn(

AdaBoostRegressor(learning_rate=0.1, n_estimators=8)

AdaBoostRegressor(learning_rate=0.1, n_estimators=8)


adb_reg_hyp = AdaBoostRegressor(learning_rate=0.1, n_estimators=76)
adb_reg_hyp_model = adb_reg_hyp.fit(xtrain, ytrain)
adb_reg_hyp_model

AdaBoostRegressor(learning_rate=0.1, n_estimators=76)

AdaBoostRegressor(learning_rate=0.1, n_estimators=76)


ytrain_pred = adb_reg_hyp_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 2109379.248700918
mean absolute error : 1258.6591436938827
Root mean squared error : 1452.3702175068581
R2 score: 0.002831057162586914


ytest_pred = adb_reg_hyp_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2084796.4621296413
mean absolute error : 1251.2237551578348
Root mean squared error : 1443.882426698809
R2 score: -0.001206046556177398


gdb_reg_model = GradientBoostingRegressor().fit(xtrain, ytrain)
gdb_reg_model

GradientBoostingRegressor()

GradientBoostingRegressor()


ytrain_pred = gdb_reg_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 2001201.1718399662
mean absolute error : 1223.691900725033
Root mean squared error : 1414.6381770049775
R2 score: 0.05397018665201092


ytest_pred = gdb_reg_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2095747.1312670347
mean absolute error : 1252.2144979335435
Root mean squared error : 1447.6695518201088
R2 score: -0.006465013727963154


hyp = {"n_estimators" : np.arange(2,100),
       "learning_rate" : [0, 0.1, 0.001, 0.0001, 1]}


RandomizedSearchCV(gdb_reg_model, hyp, cv=5).fit(xtrain, ytrain).best_estimator_

GradientBoostingRegressor(learning_rate=0.001, n_estimators=37)

GradientBoostingRegressor(learning_rate=0.001, n_estimators=37)


gdb_reg_hyp_model = GradientBoostingRegressor(learning_rate=0.001, n_estimators=42).fit(xtrain, ytrain)
gdb_reg_hyp_model

GradientBoostingRegressor(learning_rate=0.001, n_estimators=42)

GradientBoostingRegressor(learning_rate=0.001, n_estimators=42)


ytrain_pred = gdb_reg_hyp_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")

mean squared error : 2114316.2139810575
mean absolute error : 1260.159652620357
Root mean squared error : 1454.0688477445135
R2 score: 0.0004972006726960965


ytest_pred = gdb_reg_hyp_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")

mean squared error : 2084210.8770516566
mean absolute error : 1251.0061098955455
Root mean squared error : 1443.6796310302561
R2 score: -0.0009248242251242988


svc_model = SVC().fit(xtrain,ytrain)
svc_model

SVC()

SVC()


ytrain_pred = svc_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")


ytest_pred = svc_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_pred)
print(f"R2 score: {r2score}")


hyp = {"kernel" : ["linear", "rbf"], "C" : np.arange(2,15)}
hyp


svc_hyp = SVC(kernel = "linear", C =7)
svc_hyp_model = svc_hyp.fit(xtrain,ytrain)
svc_hyp_model


ytrain_pred = svc_hyp_model.predict(xtrain)


mse = mean_squared_error(ytrain,ytrain_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytrain,ytrain_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytrain,ytrain_pred)
print(f"R2 score: {r2score}")


## model Evaluation for Testing


ytest_pred = svc_hyp_model.predict(xtest)


mse = mean_squared_error(ytest,ytest_pred)
print(f"mean squared error : {mse}")

mae = mean_absolute_error(ytest,ytest_pred)
print(f"mean absolute error : {mae}")

rmse = np.sqrt(mse)
print(f"Root mean squared error : {rmse}")

r2score =r2_score(ytest,ytest_p

	Type_of_Renewable_Energy	Installed_Capacity_MW	Energy_Production_MWh	Energy_Consumption_MWh	Energy_Storage_Capacity_MWh	Storage_Efficiency_Percentage	Grid_Integration_Level	Initial_Investment_USD	Funding_Sources	Financial_Incentives_USD	GHG_Emission_Reduction_tCO2e	Air_Pollution_Reduction_Index	Jobs_Created
0	4	93.423205	103853.2206	248708.4892	2953.248771	89.887562	4	4.732248e+08	1	9.207772e+06	6663.816572	81.742461	1366
1	4	590.468942	190223.0649	166104.1642	5305.174042	84.403343	4	1.670697e+08	2	1.685101e+06	30656.049820	78.139042	1743
2	1	625.951142	266023.4824	424114.6308	2620.192622	60.498249	2	8.463610e+07	2	5.111813e+06	1749.613759	8.461296	363
3	1	779.998728	487039.5296	308337.7316	1925.250307	86.897861	3	3.967690e+08	2	4.805902e+06	43233.237820	8.402441	2821
4	3	242.106837	482815.0856	360437.7705	3948.945383	70.949351	2	3.574413e+07	1	1.668601e+07	14858.662760	28.822867	2583
...	...	...	...	...	...	...	...	...	...	...	...	...	...
14995	3	745.032555	280007.5738	230544.8268	4351.687893	90.791405	4	3.484136e+08	2	1.558508e+07	25234.911810	78.923200	1452
14996	1	15.187023	377340.5803	358547.3589	6792.194696	78.252040	4	2.560179e+08	3	6.866618e+06	15762.519790	54.982974	2598
14997	3	877.539059	480497.3920	214441.6719	4588.725297	58.282928	1	1.300112e+08	2	3.837764e+06	44597.809410	43.915897	2713
14998	7	551.264716	436383.1694	137043.8713	7251.144215	73.573666	2	3.334831e+08	2	5.347706e+06	34363.858000	4.877145	2128
14999	3	863.421803	314014.3005	124461.9178	7366.166362	86.868893	4	2.989825e+08	2	1.832534e+07	27193.217600	50.559075	1871

	Type_of_Renewable_Energy	Installed_Capacity_MW	Energy_Production_MWh	Energy_Consumption_MWh	Energy_Storage_Capacity_MWh	Storage_Efficiency_Percentage	Grid_Integration_Level	Initial_Investment_USD	Funding_Sources	Financial_Incentives_USD	GHG_Emission_Reduction_tCO2e	Air_Pollution_Reduction_Index	Jobs_Created
count	15000.000000	15000.000000	15000.000000	15000.000000	15000.000000	15000.000000	15000.000000	1.500000e+04	15000.000000	1.500000e+04	15000.000000	15000.000000	15000.000000
mean	3.973933	495.855747	252350.955621	225981.854966	5030.196472	75.219334	2.501267	2.514849e+08	2.004000	1.002977e+07	25234.722158	50.724179	2502.668600
std	1.999380	288.212872	144062.915425	129223.677997	2894.096326	14.485289	1.123306	1.432843e+08	0.817493	5.787303e+06	14378.915277	28.556578	1451.212661
min	1.000000	1.091767	1030.103692	584.048006	2.200208	50.003494	1.000000	1.008107e+06	1.000000	5.155842e+04	100.974460	1.009712	10.000000
25%	2.000000	245.475737	128568.875900	115587.428225	2543.341380	62.694076	1.000000	1.259709e+08	1.000000	4.963454e+06	12754.476927	26.224426	1228.000000
50%	4.000000	492.907555	253216.871250	225226.374350	5054.036248	75.279702	3.000000	2.539910e+08	2.000000	1.002414e+07	25424.477000	50.250207	2496.000000
75%	6.000000	742.254682	377083.805575	338656.214700	7536.935405	87.754318	4.000000	3.759130e+08	3.000000	1.507417e+07	37750.426285	75.720133	3765.000000
max	7.000000	999.982979	499991.200400	449922.667800	9999.145037	99.994955	4.000000	4.999407e+08	3.000000	1.999855e+07	49997.578530	99.980494	4999.000000

	Type_of_Renewable_Energy	Installed_Capacity_MW	Energy_Production_MWh	Energy_Consumption_MWh	Energy_Storage_Capacity_MWh	Storage_Efficiency_Percentage	Grid_Integration_Level	Initial_Investment_USD	Funding_Sources	Financial_Incentives_USD	GHG_Emission_Reduction_tCO2e	Air_Pollution_Reduction_Index	Jobs_Created
Type_of_Renewable_Energy	3.997520e+00	-3.625395e-01	1.215235e+03	-3.220774e+02	1.917457e+01	-1.123573e-01	-0.014068	-4.578642e+06	-0.010630	-2.063998e+04	4.161362e+02	-4.239348e-01	4.744598e-01
Installed_Capacity_MW	-3.625395e-01	8.306666e+04	-4.782348e+05	2.477787e+05	9.589998e+03	1.914227e+00	-3.490605	2.676324e+08	-0.499094	5.681402e+06	5.716652e+04	-6.180542e+01	-5.577267e+02
Energy_Production_MWh	1.215235e+03	-4.782348e+05	2.075412e+10	-6.972231e+06	-1.747641e+06	5.013188e+02	-417.277793	-1.254754e+11	-1267.525076	1.062634e+09	-8.600616e+06	-3.206321e+04	-4.846938e+05
Energy_Consumption_MWh	-3.220774e+02	2.477787e+05	-6.972231e+06	1.669876e+10	-4.565008e+05	1.133853e+04	-397.011971	-3.983686e+11	-311.265800	7.976445e+09	1.898074e+07	1.103052e+04	1.865765e+06
Energy_Storage_Capacity_MWh	1.917457e+01	9.589998e+03	-1.747641e+06	-4.565008e+05	8.375794e+06	4.318747e+02	-24.887617	1.163868e+09	18.430238	2.370717e+08	-7.279311e+04	-5.027205e+02	4.037827e+03
Storage_Efficiency_Percentage	-1.123573e-01	1.914227e+00	5.013188e+02	1.133853e+04	4.318747e+02	2.098236e+02	0.311152	7.897642e+06	-0.049555	2.316351e+04	1.381326e+03	3.198205e-01	2.009514e+02
Grid_Integration_Level	-1.406792e-02	-3.490605e+00	-4.172778e+02	-3.970120e+02	-2.488762e+01	3.111520e-01	1.261816	-2.401866e+05	0.007995	-1.469872e+04	2.196670e+00	1.425433e-01	1.336476e+00
Initial_Investment_USD	-4.578642e+06	2.676324e+08	-1.254754e+11	-3.983686e+11	1.163868e+09	7.897642e+06	-240186.628659	2.053038e+16	-473835.832677	1.840326e+12	-2.267804e+10	-7.335563e+06	-4.683964e+08
Funding_Sources	-1.062978e-02	-4.990944e-01	-1.267525e+03	-3.112658e+02	1.843024e+01	-4.955450e-02	0.007995	-4.738358e+05	0.668295	-1.125168e+04	1.344563e+01	-5.655922e-02	1.443289e+01
Financial_Incentives_USD	-2.063998e+04	5.681402e+06	1.062634e+09	7.976445e+09	2.370717e+08	2.316351e+04	-14698.721965	1.840326e+12	-11251.676358	3.349287e+13	3.112938e+08	-6.781600e+05	-1.404093e+08
GHG_Emission_Reduction_tCO2e	4.161362e+02	5.716652e+04	-8.600616e+06	1.898074e+07	-7.279311e+04	1.381326e+03	2.196670	-2.267804e+10	13.445634	3.112938e+08	2.067532e+08	-4.218332e+03	2.538464e+04
Air_Pollution_Reduction_Index	-4.239348e-01	-6.180542e+01	-3.206321e+04	1.103052e+04	-5.027205e+02	3.198205e-01	0.142543	-7.335563e+06	-0.056559	-6.781600e+05	-4.218332e+03	8.154781e+02	1.882566e+02
Jobs_Created	4.744598e-01	-5.577267e+02	-4.846938e+05	1.865765e+06	4.037827e+03	2.009514e+02	1.336476	-4.683964e+08	14.432888	-1.404093e+08	2.538464e+04	1.882566e+02	2.106018e+06

	Type_of_Renewable_Energy	Installed_Capacity_MW	Energy_Production_MWh	Energy_Consumption_MWh	Energy_Storage_Capacity_MWh	Storage_Efficiency_Percentage	Grid_Integration_Level	Initial_Investment_USD	Funding_Sources	Financial_Incentives_USD	GHG_Emission_Reduction_tCO2e	Air_Pollution_Reduction_Index
0	4	93.423205	103853.2206	248708.4892	2953.248771	89.887562	4	4.732248e+08	1	9.207772e+06	6663.816572	81.742461
1	4	590.468942	190223.0649	166104.1642	5305.174042	84.403343	4	1.670697e+08	2	1.685101e+06	30656.049820	78.139042
2	1	625.951142	266023.4824	424114.6308	2620.192622	60.498249	2	8.463610e+07	2	5.111813e+06	1749.613759	8.461296
3	1	779.998728	487039.5296	308337.7316	1925.250307	86.897861	3	3.967690e+08	2	4.805902e+06	43233.237820	8.402441
4	3	242.106837	482815.0856	360437.7705	3948.945383	70.949351	2	3.574413e+07	1	1.668601e+07	14858.662760	28.822867
...	...	...	...	...	...	...	...	...	...	...	...	...
14995	3	745.032555	280007.5738	230544.8268	4351.687893	90.791405	4	3.484136e+08	2	1.558508e+07	25234.911810	78.923200
14996	1	15.187023	377340.5803	358547.3589	6792.194696	78.252040	4	2.560179e+08	3	6.866618e+06	15762.519790	54.982974
14997	3	877.539059	480497.3920	214441.6719	4588.725297	58.282928	1	1.300112e+08	2	3.837764e+06	44597.809410	43.915897
14998	7	551.264716	436383.1694	137043.8713	7251.144215	73.573666	2	3.334831e+08	2	5.347706e+06	34363.858000	4.877145
14999	3	863.421803	314014.3005	124461.9178	7366.166362	86.868893	4	2.989825e+08	2	1.832534e+07	27193.217600	50.559075

Jobs Preddiction by using Machine Learning¶

Import Libraries¶

Data Gathering¶

EDA - Exploratory Data Analysis¶

assimption 2 no multicoliniarity¶

train test split¶

Model Training¶

Algorithm 1 : Logistic Regression¶

Model Evaluation for trining¶

Model Evaluation Fore Testing¶

Algorithm 2 : Decision Tree¶

model Evaluation for training¶

model Evaluation for Testing¶

Decision Tree With Hyperparameter Tunning¶

model Evaluation for training¶

model Evaluation for Testing¶

Algorithm 3 : Random Forest¶

model Evaluation for training¶

model Evaluation for Testing¶

Random Forest With Hyperparameter Tunning¶

model Evaluation for training¶

model Evaluation for Testing¶

Algorithm 4: AdaBoost¶

model Evaluation for training¶

model Evaluation for Testing¶

Adaboost with hyperprameter tunning¶

model Evaluation for training¶

model Evaluation for Testing¶

Algorithm 5 : GradientBoost¶

model Evaluation for training¶

model Evaluation for Testing¶

GradientBosst with Hyperparameter Tunning¶

model Evaluation for training¶

model Evaluation for Testing¶

Algorithm 6 : Support Vector Machine (SVC)¶

model Evaluation for training¶

model Evaluation for Testing¶

SVC by hyperparameter tunning¶

model Evaluation for training¶

	independant Features	VIF
0	Type_of_Renewable_Energy	4.643886
1	Installed_Capacity_MW	3.789933
2	Energy_Production_MWh	3.866499
3	Energy_Consumption_MWh	3.881722
4	Energy_Storage_Capacity_MWh	3.860215
5	Storage_Efficiency_Percentage	17.040044
6	Grid_Integration_Level	5.548451
7	Initial_Investment_USD	3.880253
8	Funding_Sources	6.360197
9	Financial_Incentives_USD	3.833911
10	GHG_Emission_Reduction_tCO2e	3.909498
11	Air_Pollution_Reduction_Index	3.946135